News & Views | Published:

Cell division

Burning the spindle at both ends

Accurate transmission of the genome during cell division requires the physical separation of replicated chromosomes. The identities of two molecular motors needed to do the job in fruitflies are now revealed.

A dramatic event in the life of a cell is its transformation into two genetically identical progeny. This is achieved during mitosis, when an exact complement of chromosomes is partitioned to each half of the cell, just before it pinches into two1. Errors in this process can result in cell death or contribute to cancer. The mitotic spindle — the apparatus that distributes the chromosomes — has been studied for decades. Nonetheless, the molecular mechanisms underlying chromosome transport have remained elusive. On page 364 of this issue2, however, Rogers et al. describe two related motor proteins that are essential for spindle dynamics and chromosome segregation in fruitflies.

The events of cell division require dynamic elements of the cell's internal skeleton, which assemble into macromolecular structures capable of performing work. The spindle is one such structure: this highly dynamic yet ordered assemblage is composed of cytoskeletal elements known as microtubule filaments — along with many associated proteins — which form a bipolar array3 (Fig. 1). Microtubules are polymers made up of tubulin proteins, and they grow or shrink when tubulin is added or lost from their ends. These ends are termed 'plus' and 'minus' to distinguish their behaviours. Minus ends are focused at each pole of the spindle, emanating from a nucleating structure called the centrosome. Plus ends grow outwards from the poles, capturing and moving chromosomes or overlapping at the centre of the spindle. Several classes of microtubule-based motor proteins are also required for chromosome segregation. Each type of motor moves in a set direction along microtubules. Together they cross-link and sort the microtubules according to their structural polarity, and mediate chromosome interactions with the spindle1.

Figure 1: Achieving chromosome segregation in fruitfly cells.

a, In metaphase, sister chromosomes are already attached via their kinetochores to the plus ends of kinetochore fibres (bundles of microtubule filaments), in the centre of the mitotic spindle. Sisters become aligned and oriented towards opposite spindle poles, where microtubule minus ends are focused at centrosomes. b, In anaphase, chromosomes move apart along kinetochore fibres, as the microtubules are depolymerized. c, d, Two mechanisms of chromosome movement have been proposed, based on experiments in which a marked segment (green) of the kinetochore fibre is tracked. c, In the 'reeling-in' mechanism4,5, the minus end of the kinetochore fibre is depolymerized, while the chromosome maintains attachment at the plus end. d, In the 'Pac-Man' mechanism6,7, the kinetochore fibre is chewed up from the plus end; however, the chromosome remains attached and so moves polewards. Rogers et al.2 have discovered that the motor protein KLP10A is behind the reeling-in mechanism, and KLP59C is the Pac-Man.

Before embarking on mitosis, a cell duplicates its chromosomes, producing pairs of 'sister chromatids'. The members of each pair are identical. In what is known as prometaphase of mitosis, these chromatids become tethered to the spindle, such that one member of each pair is attached to a bundle of microtubules emanating from one pole, and the other member is connected to the other pole. 'Kinetochores' are the microtubule landing pads on chromosomes3. A tug-of-war ensues, driven by the addition or loss of tubulin (that is, microtubule polymerization or depolymerization) at the kinetochores. This tug-of-war results in chromosomes becoming aligned during metaphase of mitosis, such that each sister faces its final destination (Fig. 1a). Finally, anaphase begins when the glue holding sisters together is dissolved, and they take off towards opposite spindle poles (Fig. 1b).

Long-standing questions surround the molecular mechanism of kinetochore-mediated chromosome alignment and segregation. By marking a segment of the spindle microtubule lattice and watching what happens to it as sister chromatids move apart, researchers have described two phenomena in which microtubule depolymerization is coupled to chromosome separation in anaphase (Fig. 1c, d). First, the marked region moves polewards, indicating that depolymerization is occurring at the minus ends of the microtubules; this could serve to reel in the chromosomes attached at the plus ends4,5. (Interestingly, this poleward 'flux' occurs during metaphase as well, but at this time the minus-end depolymerization is balanced by plus-end polymerization.) The second observed phenomenon is the disappearance of the labelled region as the chromosome moves past it, indicating that the kinetochores are chewing up microtubule plus ends as they move to the pole — a mechanism known as Pac-Man in reference to the video game6,7.

But what are the molecules that drive these depolymerization events during anaphase? Rogers and colleagues2 provide some answers, through a functional examination of two motor proteins in cultured fruitfly cells and early fruitfly embryos. The kinesin-like proteins KLP59C and KLP10A belong to a specialized class of microtubule-based motors (the Kin I class, for 'kinesin internal catalytic domain') that do not move along microtubules but instead bind to their ends and induce depolymerization8. During anaphase, KLP59C is found at kinetochores, whereas KLP10A is predominantly at spindle poles. Rogers et al. find that inhibiting either of these proteins interferes with chromosome segregation. By monitoring the effects on spindle microtubules, they show that KLP10A is required for microtubule depolymerization at spindle poles — that is, for the 'reeling in' mechanism. Meanwhile, KLP59C is the molecular bite behind Pac-Man (Fig. 1c, d). Interfering with these motors also causes defects in chromosome positioning during metaphase, suggesting that similar mechanisms contribute to chromosome movement during alignment.

The Kin I kinesins were logical candidates for driving chromosome dynamics through microtubule depolymerization, and have been studied previously in other experimental systems. So it is somewhat surprising that a definitive role for these proteins in anaphase has not been demonstrated before now. One explanation might be that fruitfly embryos have not been used before to study the contribution of these proteins — yet fruitfly embryos provide a unique and powerful system with which to view spindle dynamics. Their relatively large size permits analysis by high-resolution microscopy, and their rapid cell cycle does not include the quality-control 'checkpoints' found at later stages of fruitfly development and in other organisms, whereby mitotic progression is delayed if there are defects in chromosome alignment during metaphase. So, when Rogers et al. deactivated KLP10A and KLP59C, the chromosomes lined up abnormally — but compensatory mechanisms were not engaged, and effects on anaphase could still be assessed.

Another possible explanation is that members of the Kin I family also control other aspects of microtubule dynamics. So inhibiting them can give rise to marked structural defects in the spindle, producing indirect effects on chromosome movements and making it difficult to identify a specific anaphase role for the motors. And, interestingly, although functional characterization of a vertebrate Kin I relative revealed anaphase defects9, detailed analysis indicated that its primary role at the kinetochore is to depolymerize incorrectly attached microtubules, thus preventing the aberrant connections that cause lagging chromosomes and mis-segregation10. As multicellular organisms that are more complex than fruitflies possess multiple Kin-I-related proteins, it could be that the true functional counterparts of the anaphase fruitfly motors have yet to be characterized.

What remains to be discovered? One question is how microtubules maintain their attachment to kinetochores and spindle poles while undergoing polymerization and depolymerization. Also, what do motile motors at the kinetochore — such as dynein and CENP-E — contribute to the process? How are the multiple microtubules that attach to a single chromatid coordinately regulated? Mitosis provides a rich source of questions for the mechanistically inquisitive.


  1. 1

    Scholey, J. M. et al. Nature 422, 746–752 (2003).

  2. 2

    Rogers, G. et al. Nature 427, 364–370 (2004).

  3. 3

    Rieder, C. L. & Salmon, E. D. Trends Cell Biol. 8, 310–318 (1998).

  4. 4

    Mitchison, T. J. J. Cell Biol. 109, 637–652 (1989).

  5. 5

    Brust-Mascher, I. & Scholey, J. M. Mol. Cell. Biol. 13, 3967–3975 (2002).

  6. 6

    Mitchison, T. J., Evans, L., Schulze, E. & Kirschner, M. Cell 45, 515–527 (1986).

  7. 7

    Gorbsky, G. J., Sammak, P. J. & Borisy, G. G. J. Cell Biol. 104, 9–18 (1987).

  8. 8

    Desai, A., Verma, S., Mitchison, T. J. & Walczak, C. E. Cell 96, 69–78 (1999).

  9. 9

    Maney, T. et al. J. Cell Biol. 142, 787–801 (1998).

  10. 10

    Kline-Smith, S. L., Khodjakov, A., Hergert, P. & Walczak, C. E. Mol. Biol. Cell doi:10.1091/mbc.E03-08-0581 (2003).

Download references

Author information

Rights and permissions

Reprints and Permissions

About this article

Further reading


By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.